screen resistor in 5F2A

[tubeswell]

Hi there

Being the novice that I am at building/designing amps I would be grateful for some advice/education about designing my 5F2A build. I am contemplating adding a screen grid resistor to prevent the output tube plate from dissipating to much power when the amp is being overdriven. I have, on pretty good authority, a local benevolent tech (who has been very helpful in supplying mustard caps and good NOS tubes tec), that a 10k 5W screen grid resistor with a 10uF 450V cap from the screen grid pin of the 6V6GT to ground will do the job. (Actually I am using one of those new 6V6GT "tung sol" octals, so I will venture an opinion of this when the thing is all fired up).

The power supply is a(n otherwise) standard 5F1 power supply with slight changes in values (20uF for the reservoir cap and 22uF for the next two decoupling caps. There is no choke and I am using 2 x 4k7 10W resistors in series, in lieu of the first supply (aka screen node?*) resistor - 10k 1W - in the original 5F1).

However, after looking at the design on Steve Milberger's DYI Princeton, he uses a 5k2W supply (aka the screen node?*) resistor between the first two filter caps and a 1k5 2W screen grid resistor and doesn't have any decoupling cap from the screen grid pin.

For Steve Milberger's schematic See:

http://www.diycustomamps.com/images/...eton_schem.pdf

Original 5F2A schematic:

http://www.diycustomamps.com/images/...f2-a_schem.gif

Original 5F1 Schematic (for other ref to power supply):

http://www.diycustomamps.com/images/..._5f1_schem.gif


I am taking a wild guess that the lower value screen grid resistor in Steve Milberger's design means both that:

a) the voltages with not be upset too much and there is no need for the extra decoupling cap. Whereas the way my good tech has advised to build it, there are no other changes to the power supply, so the larger 10k screen resistor would otherwise upset the voltage supply section more. (Please somebody correct me if I'm wrong here?); and

b) the voltages on the screen and plate in Steve Milberger's design are a lot closer to each other, whereas I am thinking that maybe the way my good tech is recommending, that the screen voltage is effectively being knocked-for-six, (or kicked-for-touch*), and that the output valve would therefore be functioning closer to triode mode that pentode mode? (am I wayyyyy off-beam here???).

I note in another discussions on this forum, that Bruce from Mission amps has recommended keeping the screen and plate voltages within a few volts of each other for quote "a happy tube life and a rockin' dog tone".

So if I put in a 10k screen grid resistor, while I might have a happy tube life, will I still have a rockin' dog tone? Or should I build it Steve Milberger's way? and If I do build it his way, can I get away with a 20uF reservoir cap off the recto tube (a 5Y3GT) or do I need to go to a 40 or 50uF?

I don't want to lose the class A/champ characteristics of the amp. Am I being too paranoid about overheating the plate? and should I just not bother about having any screen grid attentuation (like the original 5F2A)? Am I being too precious and will the tube handle it?

Link to tube spec: ... (Hang on a minute).....(well I was lookin' hard for a link to the Tung Sol 6V6Gt tube spec , but they don't seem to have this bit of the new-sensor site working today ;-0 ), but they ae supposedly designed to "safely handle the higher voltages used in guitar amps – plus heavier plate and grid materials". Without the tube specs I can't tell what they mean by that (Couldn't find it on Duncan's pages either)

So the new-sensor Tung Sol 6V6GT might possibly be able to go into a 5F2A without all this bother???

Comments/answers/responses welcomed please.

Does my going-round-in-circles drive anybody mad?

* = Pardon my ignorance, but is this the correct American terminology?

[MWJB]

Build it stock (as per the Fender schem). Millburger's design either uses a lower voltage B+ secondary winding, or he wants to increase preamp voltage for more chime. A smaller value between the main B+ node and the screen supply reduces distortion.

An additional screen grid resistor from the screen supply on the B+ rail (point B on Millberger's schem?) should not be necessary, even if you wanted to fit one 470ohms to 1K would be in the normal range, another 10K resistor between the B+ rail & pin 4 would be overkill.

Another filter cap from 6V6 pin 4 is unnecesary.

A 40-50uf cap at the first B+ node (straight after the rectifier) will be fine.

Many amps designed to run 6V6s in the '60s, ran them at their limits. Today wall AC voltages are higher so current manufacturers build their tubes with more voltage handling capacity so that they don't go pop the minute someone puts them in a BF/SF Deluxe or Princeton.

Your amp should not run voltages as high as these amps if you use a typical 125P1B power transformer and a NOS 5Y3GT.

[tubeswell]

Thanks MWJB.

In reponse to your comment on the size of the filter cap, I was intending to build it using a 20uF 500V Sprague Atom for the first filter cap. I am wanting a typical champy-sort-of-amp, but with a tone control (which I thought was sort of what the 5F2A was, albeit that the Fender schematic has two 16uF in parallel for the first filter.). I have 22uF 500V for the 2nd and 3rd caps in the power supply.

Do you still see no issue with the screen grid vs Plate if I do it this way?

Cheers

[MWJB]

As long as the resistor between points A & B on Millberger's schematic is more than 2 or 3K you shouldn't have any issues. With the stock 10K you should be laughing.

Tweed Champs and Princetons used smaller caps than 20uf at preamp & screen supplies, these will have more impact on your tone than the first filter cap. That said, if you have 22uf caps try them it won't sound bad. I'd go bigger at the first filter cap just to help with hum, can you fit 2x20uf in parallel?

[tubeswell]

Thanks MWJB


My PT (that I had to get specially made up) has a CT on the Heater winding, which I guess I could run to the cathode pin on the 6V6GT as a next resort. With the snow-flake grounding scheme I have in mind, (Pre-amps together before joining with rest of amp grounds in one spot) I've got my fingers crossed that hum won't be too bad. So I'll try it with one 20uF at the first cap, and add a parallel one if its too hummy.

[MWJB]

"With the snow-flake grounding scheme I have in mind, (Pre-amps together before joining with rest of amp grounds in one spot)" I really wouldn't do this. Ground the main filter & screen filter at a PT bolt, ground everything (cathodes, preamp filter) else at #1 input jack, separate wire from each eyelet to the ground point (not daisy chained together, no accidental ground points at points other than prescribed grounds).

Yes, you could try DC biasing the heaters via the 6V6 cathode. It is usually the heaters that cause hum in SE amps. You could also try DC biasing the heaters via a voltage divider from the first filter cap (search the forum for this there are important points to bear in mind).

The only reason Fender used smaller caps at the first filter was for reasons of cost. I'd go stright in with 40uf. If your amp does hum there might be multiple causes if this is your first build, so why not eliminate what possible causes that you can now?

[tubeswell]

Grounding plan for 5F2A build

"Ground the main filter & screen filter at a PT bolt, ground everything (cathodes, preamp filter) else at #1 input jack, separate wire from each eyelet to the ground point (not daisy chained together, no accidental ground points at points other than prescribed grounds).

Thanks for the tips on grounding MWJB. Is this what you meant, or should I have separate ground wires for each of the first and second filter caps as well?

(note that the output sockets are separately grounded - I take it that's okay? Please say if not.)

Also would appreciate any comments you have on the text I've highlighted in red thanks.

Cheers

[MWJB]

470ohm cathode resistor, screen supply filter cap & main filter cap should have their own wires to a PT bolt. (It doesn't usually make a great deal of difference in this design whether cathode resistor gets grounded to PT bolt or buss wire but as you've isolated the 470ohm grounded eyelet you may as well go for the PT bolt).

Easy enough to ground the AC cord to the bottom left (or top left) PT bolt, power supply grounds will be easier to ground top one of the RH PT bolts.

Are you using a bus wire or brass grounding plate for the input jacks & pots? Vol pot LH tab gan be grounded to pot body as can 0.0047uf cap on tone pot. This will make life easier too as you will only have 3 wires, plus buss wire, going to input#1 ground. The buss wire will also give a convenient point to ground your sheilded cable to V1, pin 2.

If all is laid out well you will not need grid stopper resistor or screen grid (6V6 pin 4 & 5).

OT will be OK mounted vertically on the chassis (stand up) if the PT is the "lay down/drop through" type, as laminates will be at 90degrees to each other anyway.

OK to ground common OT secondary to output jack grounds.

[tubeswell]

470ohm cathode resistor, screen supply filter cap & main filter cap should have their own wires to a PT bolt.

- So does this mean "don't ground these with the PT grounds"?.

Easy enough to ground the AC cord to the bottom left (or top left) PT bolt, power supply grounds will be easier to ground top one of the RH PT bolts.

- I have grounded the PT Shield, the AC mains earth, and the Heater CT, and the HT winding CT (via a standby switch) to one of the PT bolts. (I was planning to ground the 470ohm cathode resistor, screen supply filter cap & main filter cap to the same PT bolt, but I will change that plan if your answer to the above question is 'yes'.)

Are you using a bus wire or brass grounding plate for the input jacks & pots?

- Well I was going to use a bus wire system, but I thought you suggested each preamp ground-side part should have a separate wire to the No.1 input ground lug. When I tried this, it wasn't big enough for 7 ground wires (including the shield for the V1A grid wire), so I mounted all these ground leads on a terminal post (like a star ground), which only has one wire going to the No.1 Ground Lug. (Maybe I was confused about this - at the moment its not too much hassle to undo, or if its okay to leave it the way it is I will).


If all is laid out well you will not need grid stopper resistor or screen grid (6V6 pin 4 & 5).

- Okay cool

OT will be OK mounted vertically on the chassis (stand up) if the PT is the "lay down/drop through" type, as laminates will be at 90degrees to each other anyway.

- I have the PT and OT laminates mounted at right angles to each other, and at opposite ends of the chassis (about 1 foot apart)

OK to ground common OT secondary to output jack grounds.

Cheers

[MWJB]

470ohm , main & screen filters can be grounded to the same point as PT grounds, they just require their own wire from each eyelet on the board (not daisy chained to each other, then share a wire to ground). Yes, they can got to the same PT bolt as CT's/PT shield etc). Mains AC ground may be better off with it's own PT bolt?

"Well I was going to use a bus wire system, but I thought you suggested each preamp ground-side part should have a separate wire to the No.1 input ground lug. When I tried this, it wasn't big enough for 7 ground wires (including the shield for the V1A grid wire), so I mounted all these ground leads on a terminal post (like a star ground), which only has one wire going to the No.1 Ground Lug. (Maybe I was confused about this - at the moment its not too much hassle to undo, or if its okay to leave it the way it is I will)." They way you have things should be OK, as long as the wire from the terminal post to #1 jack ground is very short and a good heavy guage (we are trying to eliminate the different circuit grounds sharing the same wire to a ground point). I usually tie the preamp grounds to the buss wire, right where it meets the #1 jack ground lug, with pot grounds soldered to the pot body this is only usually a couple of wires, plus shielded cable ground.

"I have the PT and OT laminates mounted at right angles to each other, and at opposite ends of the chassis (about 1 foot apart)" that'll probably be OK for this kind of design, however, keeping the OT away from the input jacks is a good idea, especially with higher gain designs.

[tubeswell]

That clears a heap of things up. Cheers and Thanks for all your help so far MWJB.

[Paul P]

I'm curious why you need the equivalent of a 10k ohm 20w resistor
between the OT and the power tube screens. Wouldn't 2w be enough ?
Fender used 1w's.

Paul P

[Bruce / Mission Amps]

I'm curious why you need the equivalent of a 10k ohm 20w resistor
between the OT and the power tube screens. Wouldn't 2w be enough ?
Fender used 1w's.

Paul P

Yes a 1 to 2 watt screen resistor is plenty good enough for 6V6s, 6L6s, KT66... I'd use a 5 watter for EL34s as a precaution.

[tubeswell]

I'm curious why you need the equivalent of a 10k ohm 20w resistor
between the OT and the power tube screens. Wouldn't 2w be enough ?
Fender used 1w's.

Paul P

Yep I saw that when I was first doing the design, but someone (I forgot who) said that the 1W resistors didn't cope all that well with the amp being driven hard and it would be safe to stick in a 10W resistor in that location. I couldn't find a 10k 10W locally, but I found plenty of 4k7s at 10W. So there you go - its a bit of a stock design with a few extra features ;-)